Re: bugzilla abuse

[Tim Janik <timj gtk org>]

On 5 Jan 2001, Havoc Pennington wrote:

> 
> Tim Janik <timj gtk org> writes:
> > 
> > you recently submitted a bunch of bogus gobject bugs to bugzilla.

> The point is to get all possible issues tracked for the 2.0
> release. If we decide things are not issues, then we close the
> bug. Bugs are *POSSIBLE* issues that must be resolved before
> release. Resolution can be "NOTABUG", of course, if it turns out to be
> a non-bug.

as i mailed to owen already, that requires the system to be readily
accesible to everyone involved, and even then, i don't want to spend
all day cleaning up the bug tracker from issues that you made up or
had already been adressed i nthreads.

> > 2) some of your bugs had already been adressed in previous threads, bugzilla
> >    is not your personal wishlist bag for "i still want things the xxx way done"
> >    things
> 
> Many of those bugs are things that I at least asked Owen about
> beforehand, or sometimes Jonathan or others, so they are all issues
> that can't be considered finally resolved and need discussion. That's
> all. If you consider some of them already discussed and closed issues,
> then I don't know which ones; and I have no way of knowing which
> ones. You have to tell us what the hell is going on. We are not over
> here receiving Tim brain waves from Germany via ESP.

lucky me that i do receive brain waves from all the rest of lab rats
all day long? ;-O

> A bug means "this is a possible issue we aren't sure is resolved."
> It's not some kind of personal insult... just a request to be sure
> everyone is happy the issue is resolved. There are a ton of bugs
> applying to everyone's code, not just GObject. So you should not take
> this personally or anything.

i certainly didn't take that personally until your brain wave commentary.

> > 3) it doesn't help any to submit things as bug reports just because they're
> >    currently being worked on but haven't ended up in CVS yet
> 
> Yes it does. We need to know which issues are outstanding. Otherwise
> we don't know when we are finished or what needs doing. The rest of us
> have no idea what is on your hard drive. Bugs are fixed when the code
> is in CVS, not before; because the point of tracking issues is to be
> sure everything makes it into the final release.
> 
> You should perhaps think of bugzilla as an "issue tracker" not a bug
> tracker.

once more, that's not a sufficient solution until we can all use bugzilla
to its fullest extend, not having an email interface means you lack a rough
150 lines of issue list that still sits only on my local harddisk.

> > last but not least, i personally am not going to use bugzilla seriously
> > anyways, until there's a decent and functionally equivalent email interface
> > to it. maintenance where every bug evaluation or commenting thereof
> > requires me to be online is simply not an option.
> > 
> 
> You can proxy through me and Owen or gtk-devel-list, I'm happy to
> close bugs or add links to gtk-devel-list posts in the comment section
> on your behalf. i.e. if you just tell us what is going on, we are
> using bugzilla daily to keep track of what's going on, so we can make
> bugzilla reflect what you tell us.

instead of having you track my emails/commits, i'd rather see you (or anyone
else for that matter) spend that extra time on an email interface to bugzilla
so non-24/7 type of guys have something to work with as well.

> 
> Havoc
> 

---
ciaoTJ